岩石物理驱动的储层物性参数非线性地震反演方法

叠前地震反演和岩石物理反演分别是获取弹性参数和物性参数的重要手段, 两者结合有助于实现储层参数预测并精细刻画储层特征.储层物性参数的反演依赖于岩石物理模型, 在进行物性参数反演时可以将复杂的岩石物理模型做泰勒展开, 进而得到其一阶或高阶的近似表达式, 然而这会降低模型的精确性并增加反演的误差.为了提高储层物性参数反演的稳定性和准确性, 本文以碎屑岩储层为例, 提出了岩石物理驱动的储层物性参数非线性地震反演方法.首先, 基于贝叶斯框架和高斯分布约束条件, 从叠前地震数据中实现纵、横波速度及密度等弹性参数的反演.其次, 通过碎屑岩岩石物理模型建立起弹性参数与物性参数之间的联系.最后, 利用粒子群算法进行全局寻优获得较为准确的孔隙度、泥质含量和含水饱和度等物性参数.合成数据和实际资料测试结果验证了所提方法的可行性和准确性, 反演结果与测井数据吻合较好, 可有效指示含气储层区域, 本文方法在储层预测和评价方面具有广泛的应用前景.

Rock-physics-driven nonlinear seismic inversion for petrophysical parameters of reservoir

Pre-stack seismic inversion and petrophysical inversion are important tools to estimate the elastic and petrophysical parameters, and the combination of the two is helpful to predict reservoir parameters and characterize reservoir features more precisely. The inversion of petrophysical parameters of reservoir relies on rock-physics models. In petrophysical parameters inversion, complex rock-physics models can be Taylor expanded to obtain their first-order or higher-order approximate expressions, however, this can reduce the accuracy of the models and increases the inversion error. To improve the stability and accuracy of petrophysical parameters inversion, this paper proposes a rock-physics-driven nonlinear seismic inversion for petrophysical parameters of reservoir, taking clastic reservoirs as an example. Firstly, the estimations of elastic parameters such as P- and S-wave velocities and density are realized from pre-stack seismic data based on the Bayesian framework and Gaussian distribution constraints. Secondly, the connection between the elastic parameters and the petrophysical parameters is established through the clastic rock-physics model. Finally, the particle swarm algorithm is used to accurately estimate petrophysical parameters such as porosity, clay volume and water saturation by global optimization. The synthetic and field data test results verify the feasibility and accuracy of the proposed method, and the inversion results are in good agreement with the logging data and can effectively indicate the gas-bearing reservoir area. The method in this paper holds broad application prospects in reservoir prediction and evaluation.